The performance of the vestibular system can be quantified by monitoring the motion of the eyes due to the vestibular-ocular reflex (VOR) as it is being activated by applying a rotation motion to the experimental subject or by directly stimulating the semicircular canals. Our laboratory developed a vestibular prosthesis. This device contains gyroscopes that can sense motion about three axes of rotation and it has the ability to provide electrical stimulation pulses to the electrodes, attached to the device. The initial tests of this device with the electrodes implanted in the semicircular canals of chinchillas indicate that the separation between the branches of the vestibular nerve is too narrow to allow a clean stimulation signal to be delivered separately to each targeted branch of the vestibular nerve. The goal of this project is to establish the electrical stimulation paradigm which would provide a meaningful set of vestibular signals to the brain despite the signal crosstalk due to non-specific stimulation. We will examine two possible solutions to the problem of electrode interaction and adapt a spike integrator model to predict the effect of stimulus train parameters on the VOR response. Aim 1) Linear relationship between electrical stimulation and VOR response would allow us to use well- established 3D rotational kinematics to process the gyroscope signals from the prosthesis, despite electrode interaction. We hypothesize that linear relationships hold true between electrical stimulation and VOR responses. Aim 2) Low amplitude stimulation improves electrode specificity, but decreases the strength of the VOR response. High stimulation rates have been shown to introduce strong VOR response. We hypothesize that using low amplitude stimulation pulses delivered at high stimulation modulation rates will produce strong VOR responses with low electrode interaction. Aim 3) Finally, we will extend a biophysical model developed for predicting perceived intensity of stimulation pulse trains delivered to the somatosensory cortex, for the purposes of vestibular stimulation. We hypothesize that the model will provide accurate representation of VOR response to a wide variety of electrical stimulation parameters. RELEVANCE: Bilateral dysfunction of the vestibular system has debilitating effects on the patients, who are forced into a slow motion existence so as to compensate for the lack of their balance system. This study examines several techniques to enable and enhance the functioning of an implantable vestibular prosthesis.